[Paper Review] A grid of MARCS model atmospheres for late-type stars I. Methods and general properties

B. Gustafsson, B. Edvardsson+4 more|ArXiv.org|May 5, 2008

Stellar, planetary, and galactic studiesPhysics and Astronomy87 references1,267 citations

TL;DR

This paper presents a comprehensive grid of 10,000 spherically symmetric MARCS model atmospheres for late-type stars, covering a wide range of effective temperatures (2500–8000 K), surface gravities (log g = –1 to 5), metallicities ([Me/H] = –5 to +1), and abundance variations including C/N and C/O ratios. The models incorporate up-to-date atomic and molecular data, LTE, mixing-length convection, and hydrostatic equilibrium, and show excellent agreement with Kurucz and PHOENIX grids, revealing complex feedbacks between sphericity and molecular blanketing in cool stars.

ABSTRACT

We have constructed a grid of about 10,000 spherically symmetric and plane-parallel models with the MARCS program, and make it available for public use. Parameter ranges are: Teff=2500 to 8000 K, log g =log(GM/R2)= -1 to 5 (cgs) with various masses and radii, [Me/H]=-5 to +1, with [Alpha/Fe] = 0.0 and 0.4 and different choices of C and N abundances to also represent stars of types R, S and N, and with microturbulence parameters from 1 to 5 km/s. We also list fluxes in approximately 108,000 wavelength points. Underlying assumptions in addition to 1D stratification include hydrostatic equilibrium, MLT convection and LTE. A number of general properties of the models are discussed, in relation to the effects of changing blanketing and sphericity. Models are compared with other available grids and excellent agreement is found with plane-parallel models of Castelli and Kurucz within the overlapping parameter range. Although there are departures from the spherically symmetric NextGen models, the agreement with more recent PHOENIX models is gratifying. The models of the grid show regularities, but some interesting departures from general patterns occur for the coolest models due to the molecular opacities. We have tested rules of thumb concerning effects of blanketing and sphericity and found them to often be astonishingly accurate. Some interesting new phenomena have been discovered, such as the intricate coupling between blanketing and sphericity, and the strong effects of carbon enhancement on metal-poor models. We give further details of models and comparisons with observations in subsequent papers.

Motivation & Objective

• To develop a homogeneous, publicly available grid of model atmospheres for late-type stars to support spectral and photometric analysis.
• To improve modeling accuracy by incorporating state-of-the-art atomic and molecular line data, including blanketing effects.
• To explore the impact of sphericity, metallicity, and carbon enhancement on model structure and fluxes.
• To ensure consistency and interoperability with existing grids like Kurucz and PHOENIX through rigorous comparison.
• To provide interpolation tools and detailed thermodynamic data for use in stellar atmosphere and abundance analysis.

Proposed method

• The MARCS code is used to compute spherically symmetric model atmospheres, with plane-parallel models for high surface gravity.
• The models are constructed under assumptions of 1D stratification, hydrostatic equilibrium, mixing-length convection, and LTE.
• Atomic and molecular line data are updated using VALD, OP, and IP databases, with extensive line lists for molecules and ions.
• Thermodynamic quantities (T, P_g, P_e, ρ, partial pressures) and Rosseland mean opacity are calculated at ~108,000 wavelength points.
• Interpolation routines are implemented to enable model access at non-tabulated parameters, validated with <0.25% error in T and a few percent in P_g and P_e.
• Comparisons are made with Kurucz (plane-parallel) and PHOENIX (3D-like) models to validate consistency and accuracy.

Experimental results

Research questions

• RQ1 How do sphericity and molecular blanketing interact in cool, low-gravity stars, and what feedback mechanisms emerge?
• RQ2 To what extent do approximate analytical rules for blanketing and sphericity hold in realistic model grids?
• RQ3 How do variations in C/N and C/O ratios affect model structure and fluxes, especially in metal-poor stars?
• RQ4 How well do MARCS models agree with existing grids such as Kurucz and PHOENIX across overlapping parameter spaces?
• RQ5 What are the quantitative impacts of metallicity, α-enhancement, and microturbulent velocity on model atmospheres and flux spectra?

Key findings

• The grid comprises approximately 10,000 model atmospheres covering T_eff from 2500 K to 8000 K, log g from –1 to 5, [Me/H] from –5 to +1, and [α/Fe] = 0.0 and 0.4.
• The models show excellent agreement with plane-parallel Kurucz models when convection is treated consistently, validating the numerical approach.
• Agreement with the more recent PHOENIX grid is gratifying, despite differences in numerical methods and data selection.
• Strong feedbacks between sphericity and molecular blanketing are identified, particularly in the coolest models, where molecular opacities dominate.
• Approximate rules of thumb for blanketing and sphericity effects are found to be astonishingly accurate across the grid.
• Carbon enhancement has a strong, non-linear effect on metal-poor models, significantly altering structure and fluxes, especially in C-rich (R, S, N-type) stars.